Weathering Steel in Industrial-Marine-Urban Environment: Field Study

P. Dhaiveegan1, N. Elangovan2, T. Nishimura3 and N. Rajendran1,+

1Department of Chemistry, Anna University, Chennai-600025, India2Department of Chemistry, A. M. Jain College, Chennai-600114, India3Material Recycling Design Group, Research Center for Strategic Materials,National Institute for Materials Science, Tsukuba 305-047, Japan

In the present field study, we report the exposure of weathering steels in the industrial-marine-urban environment at Ennore, located nearthe east coast of India for 3 years. The Corrosion products viz., iron oxyhydroxides and oxides present in the rust layers were characterized usingATR-FTIR, XRD studies and quantified using TGA analysis. The specific surface area of the rust particles formed during the corrosion processwere determined using N2 adsorption isotherm studies. The morphology of the corrosion products were elucidated using SEM. ATR-FTIR andXRD studies showed that the corrosion products formed on the skyward surfaces were highly more crystalline than those on the the earthwardsurfaces. TGA showed that the iron oxyhydroxides were major corrosion products. N2 adsorption-desorption studies confirmed the formation ofcompact inner rust layers with high specific surface area (SA). SEM analysis revealed that the skyward surface was smooth and compact whilethe earthward surface were cracked and porous. SEM images confirmed the formation of characteristic morphological structures such as sandycrystal (lepidocrocite), needle-like (goethite), cotton ball (goethite) and cigar shaped (akaganeite) structures. Further, it revealed the formation ofgoethite on the skyward surface as a major constituent phase in the rust layer, which was influenced by the presence of SO2 content in theenvironment. [doi:10.2320/matertrans.M2015345]

(Received September 7, 2015; Accepted November 12, 2015; Published January 18, 2016)

Keywords: weathering steel, X-ray diffraction (XRD), scanning electron microscope (SEM), brunauer-emmet-teller (BET), atmosphericcorrosion, rust and goethite

1. Introduction

Weathering steel is widely used as a construction materialdue to its low cost and high corrosion resistance. The highcorrosion resistance of weathering steel is attributed to theformation of dense and adherent rust layer.1) Atmosphericcorrosion studies of weathering steels in various environ-ments such as marine, urban and industrial areas werereported.2) The corrosion parameters like chloride (Cl¹),sulphur dioxide (SO2), relative humidity (RH), temperatureand composition of alloying elements play a crucial role inthe formation of rust products during the period of exposurein different environments.3­5) The characteristic nature of therust layer were well studied by several researchers over thelast few decades. 6­8)

Asami et al.9) investigated the weathering carbon steelsexposed to industrial and marine environments for 17 yearsand concluded that the distributions of ¡-FeOOH phase in theinner rust layer was uniform and served as a barrier layer inthe rust. Fuente et al.10) studied the morphology of corrosionproducts formed on mild steel, which was exposed for 13years in various environments such as marine, industrial, ruraland urban. It was found that the skyward facing surface hadsmooth and uniform structure with low aggressivity, whereasthe earthward facing surface had irregular cracked structure.Thermogravimetric analysis were used to identify andquantify the composition of the rust and this can be achievedwith a small amount of sample.11) Yuxi et al.12) quantitativelyanalysed the composition of the rust formed in variousenvironments using thermal analysis. Ishikawa et al.13)

assessed the specific surface area of the rust layer by N2 andH2O adsorption studies and related this to the extent ofcorrosion and type of steel exposed. They showed that

compact rust layers with high SA or small particle size exhibita high corrosion resistance. Singh et al.14) have studied thelong term corrosion behaviour of low alloy and plain carbonstructural steel exposed to different climatic conditions for 2years and reported the nature of rust formed, which revealedthat the presence of SO2 played a vital role in reducing thecorrosion rate at Chennai site when compared to other sites.

Ennore is situated in the northern part of Chennai city atthe east coast of South India near Bay of Bengal. Thermalpower plants, petrochemical plants, ports, refineries, pharma-ceutical companies and residential buildings are situated inEnnore and it is a marine-industrial-urban environment.There are no reports on the detailed field study of rust layersformed on weathering steels subjected to long term exposurein an industrial-marine-urban environment in India. In thepresent investigation, thermo gravimetric analysis and N2

adsorption studies were used to understand the nature ofcorrosion products formed on the skyward and earthward rustlayers of weathering steel exposed at Ennore, Chennai, India.

2. Experimental Procedure

2.1 Preparation and exposure of specimenThe composition of the exposed weathering steel is

presented in Table 1. The chemical composition of alloyingelements was confirmed by inductively coupled plasmaoptical emission spectroscopy (ICP-OES). The specimendimensions are 100mm © 50mm © 3mm. The specimenswere abraded with 800 # SiC paper, rinsed with water andacetone, dried, weighed and kept in a moisture free

Table 1 Chemical composition of the weathering steel (mass%).

Elements C Si Mn P S Cu Cr Ni Fe

Composition 0.08 0.25 0.33 0.07 0.01 0.44 0.51 0.25 Balance+Corresponding author, E-mail: nrajendran@annauniv.edu

Materials Transactions, Vol. 57, No. 2 (2016) pp. 148 to 155©2016 The Japan Institute of Metals and Materials

desiccators for 3 days prior to exposure. ASTM G 50-9715)

standard was followed and the specimens were exposed toenvironment for 36 months (January 2009-Dec 2012). Theexposure distance was 100m from the marine and 200mfrom the industrial environment. The topography map of theexposing the site is given in Fig. 1.

The climatic and environmental parameters viz., relativehumidity, temperature were obtained from Regional Mete-orological Centre, Chennai. The amounts of atmosphericdeposition of Cl¹ and SO2 content in the field were estimatedusing wet candle and sulphonation plate procedures pre-scribed in ASTM G 140-9616) and ASTM G 91-9717) werefollowed. The 3 years average environmental parameters arelisted in Table 2.

2.2 Rust separation and its analysisThe easily peelable rust layers from the skyward and

earthward facing surfaces were collected and labelled asskyward outer rust layer (SOL) and earthward outer rust layer(EOL). The adherent rust layers were carefully scrapped fromthe specimen using a sharp cut knife and were designatedsimilarly as skyward inner rust layer (SIL) and earthwardinner rust layer (EIL). The collected rust was ground and keptin a desiccator for 3 days before analysis. The attenuated totalreflection fourier transform infrared (ATR-FTIR) spectra ofrusts were recorded in the 2000­400 cm¹1 range on a PerkinElmer Spectrum Two instrument with UATR as accessoryand KBr window. A maximum of 200 scans wereaccumulated for each sample. X-ray diffractograms (XRD)were recorded using a PANalytical X’Pert PRO diffractom-eter in the 2ª range from 10° to 70° at a step size of 5°/minusing CuK¡ (­ = 0.154 nm) as X-ray Source The micro-structure and elemental distributions in the rust layers wereanalysed using a Quanta 200 FEG Scanning ElectronMicroscope equipped with energy dispersive X-ray analysis(SEM-EDAX). Thermo gravimetric analysis (TGA) of the

rust were carried out using a TA SDT Q600 thermal analyserat a heating rate of 5°C/min from room temperature to1000°C in an air atmosphere. The specific surface area SA ofthe rust were determined using N2 adsorption/desorptionstudies with quanta chrome system at 77K. Before theadsorption studies, the samples were degassed with N2 at100°C for 12 hours. The SA and pore size distributions werecalculated by using Brunauer-Emmet-Teller (BET) andBarret-Joyner-Halenda (BJH) methods, respectively.

3. Results and Discussion

3.1 Corrosive agents and environmental characteristicsin the exposure site

The ISO 9223-2012 classification can be used to predictthe nature of aggressiveness environment by determining theatmospheric pollutant parameters viz., SO2 and Cl¹. In ourcase study the atmospheric corrosion exposure location iscloser to the industrial environment, the SO2 deposits areobserved in higher amount compared to the Cl¹. Thedetermined values of Cl¹ and SO2 by wet candle andsulphonation plate confirmed the presence of SO2 (85mg/cm2/day) (P3) in the higher amount when compared toCl¹ (62mg/cm2/day) (S2). Based on the measured valuesof SO2 and Cl¹ at our location, our site can be classified asC5 with reference to ISO 9223-2012 classification.18) Theaverage temperature (36°C) and RH (98%) remain through-out the year with high values. The temperature and RH variedfrom 30°C to 39°C, 60 to 98% respectively throughout the 3years of exposure period. The major rainfall was observed inthe months of June to December during the southwest andnortheast monsoon period.

3.2 Characterization of rust layers and their products3.2.1 Surface appearance of WS after 3 years of

exposureThe physical appearance of WS surface of the skyward

(Fig. 2(a)) rust layer was light red brown, adherent, compactand more uniform, whereas the earthward (Fig. 2(b)) surfaceof the specimens was dark brown and uneven. Also thesurface was rougher and high porous in nature with the looserrust particles in EOL; but in the SOL the surface wassmoother with less porous nature.3.2.2 ATR-FTIR spectroscopic studies

The skyward (Fig. 3(a)) and earthward (Fig. 3(b)) facingrust layers were analysed using ATR-FTIR. All the ATR-FTIR spectra display the characteristic absorption peaksfor water insoluble corrosion rust products namely lepidoc-rocite (£-FeOOH-1020 cm¹1), goethite (¡-FeOOH-890 and790 cm¹1), (akaganeite) (¢-FeOOH-680 cm¹1) maghemite(£-Fe2O3-560 cm¹1).19) Water-soluble iron chloride andsulphate as residual concentration in the rust layers are hardto be detected.20) In all the spectra, the absorption peak

Fig. 1 Topography of the location of atmospheric corrosion exposure siteof Chennai, India.

Table 2 Environmental conditions of exposure site (yearly average values for the period 2009­2012).

Test site EnvironmentTemperature

(T/K)Relative humidity

(%)Cl¹ deposition(mg/cm2/day)

SO2 deposition(mg/cm2/day)

Ennore Industrial-Marine-Urban 307 98 62 85

Weathering Steel in Industrial-Marine-Urban Environment: Field Study 149

around 1620 cm¹1 corresponds to physisorbed water. Thedifference in the band intensities between 890 cm¹1 and1020 cm¹1 in the skyward rust layer is attributed to theformation of goethite over lepidocrocite. The formation ofgoethite in the SOL is thermodynamically favourable atlower pH in the presence of sulphur dioxide which isobserved at 1460 cm¹1. Figures 3(a) and (b) reveals thepresence of chloride ion favours the formation of akaga-neite.21­24) The formation of stable goethite in the skywardouter rust (SOL) layer suppresses the penetration of chlorideions to the skyward inner layer (SIL) and hence less intensepeak is observed for akaganeite at 680 cm¹1 in the SIL. Theprotective ability index (PAI) of the rust is determined by thepeak intensity ratio of the ¡-FeOOH and £-FeOOH.22) Theskyward rust layers show the higher PAI value of 1.7 whereasthe earthward rust layer shows a value of 0.98. PAI valueindicates the high corrosion resistance nature of SOL whencompare to EOL.3.2.3 XRD analysis

Figures 4(a)­(d) shows the XRD patterns of the rust layers

(a) (b)

Fig. 2 Surface appearance of (a) skyward and (b) earthward specimensexposed for 3 years at industrial-marine-urban environment.

500 1000 1500 2000

EOLA

ML

G

Tran

smitt

ance

(a.u

.)

Wavemumber, υ / cm-1

G

EILA

ML

GG

500 1000 1500 2000

Tran

smitt

ance

(a. u

.)

Wavemumber, υ / cm-1

SIL

SOL

M

HM

G

A

G

L SO42-

(a) (b)

Fig. 3 ATR-FTIR spectra; (a) SOL and SIL, (b) EOL and EIL (G-Goethite, A-Akagagenite, L-Lepidocrocite, M-Maghemite,H-Hematite).

(a) (b)

(c) (d)

Fig. 4 XRD patterns of powdered rust layers of weathering steel after 3 years of exposure; (a) SOL, (b) SIL, (c) EOL and (d) EIL.

P. Dhaiveegan, N. Elangovan, T. Nishimura and N. Rajendran150

present in SOL, SIL, EOL, EIL of the exposed weatheringsteel. The presence of iron oxides viz., maghemite (£-Fe2O3),magnetite (Fe3O4) and hematite (¡-Fe2O3) and oxyhydrox-ides viz., lepidocrocite (£-FeOOH), goethite (¡-FeOOH),akaganeite (¢-FeOOH) and ferrihydrite (¤-FeOOH) withvarying intensities in all the rust layers are observed. Therelative peak intensity of goethite (2ª = 21°) in the skywardsurface was stronger than the earthward surface which isattributed to the presence of higher sulphur content whichin turn enhances the transformation of lepidocrocite intothermally stable goethite.14) In the case of EOL relative peakintensity of lepidocrocite (2ª = 14°) is much stronger thanthe skyward surface. The relative peak intensity of akaganeitein the earthward layer is lower than the skyward layer due tothe spallation of the earthward rust layers.25) Exogeneoussilica (SiO2) is observed in all the rust layers. The cationselective ability of goethite enhances the corrosion resistanceof the steel which retards penetration of chloride from theenvironment.26) From the above results, it is inferred that theskyward surface is more corrosion resistant than the earth-ward surface. These results are in agreement with the ATR-FTIR studies.

3.3 Morphology of corrosion productsScanning electron micrographs (SEM) of the skyward and

earthward surfaces of exposed weathering steel are shownin Fig. 5. The presence of voids and micro cracks aresignificantly higher in the earthward surface when ascompared to the skyward surface (Fig. 5(a) and (b)). SEM-EDAX results for both skyward and earthward surfaces arerepresented in Fig. 5(c) and (d). It is inferred from the SEM­EDAX mapping that the distribution of relative sulphurcontent in the skyward rusted surface is higher than in theearthward surface. Further, it is inferred that the chloridecontent in the skyward surface is lower than the earthward

surface. The presence of chloride influences the formation ofcracks which are evident from the difference in the skywardand earthward micrographs. Chloride ion influences theformation of flaked rust in the earthward surface whereas,SO2 plays a crucial role in the formation of less roughermorphology in the skyward surface.9,27) Figures 6 and 7show the typical morphologies of the rust products formed onthe skyward and earthward surfaces respectively after 3 yearsof exposure in the marine-urban-industrial environment. Fineplate-like structure is present in the skyward surface(Fig. 6(a) and (b)) and the crystalline globule (sandy crystals)structure is present in the earthward surface (Fig. 7(a)). Thesestructures account for the presence of lepidocrocite (£-FeOOH). The cotton ball (Fig. 6(c)) and needle-like

(a) (c)

(b) (d)

200 μm

200 μm

Fig. 5 SEM/EDAX micrographs of the rust layers on weathering steel:(a), (c) skyward surface and (b), (d) earthward surface.

(a) (c) (e)

(b) (d) (f)

Fig. 6 SEM micrograph of typical morphologies of skyward surface corrosion products of rusted weathering steel: (a) and (b)lepidocrocite, (c) and (d) goethite and (e) and (f ) akaganeite.

Weathering Steel in Industrial-Marine-Urban Environment: Field Study 151

structures (Fig. 6(d)) in the skyward surface and whiskerycotton ball structure (Fig. 7(b)) in the earthward surface canbe regarded as the structures of goethite (¡-FeOOH).23)

Akaganeite (¢-FeOOH) is present in the form of cigarshaped crystals (Fig. 6(e)) and rosette morphology(Fig. 6(f )) in the skyward surface and whiskery rod shapestructures (Fig. 7(c)) in the earthward surface.

It is accepted that rusting is a heterogeneous process andresults in different phases.27,28) The different morphologicalstructures of the iron rust corrosion products with samephases are attributed to the time of exposure.29) Based onEvans30) and Misawa’s Model,31) the corrosion processes andthe formation of rust layers are depicted in Fig. 8. In theinitial stage of rusting lepidocrocite is formed and further thepresence of sulfur dioxide favors the formation of goethitethrough solid-state transformation.3.3.1 Cross-sectional morphology of rust layers

The cross-sectional morphology of rust layers are depictedin (Fig. 9(a)­(b)) and their EDAX spectra (Fig. 9(c)­(d))show the Cl¹ and SO2 enrichment. The rust layers are wavy

and undulating with distinguishable outer and inner layers.Among the two rust layers, the corrosion products in theskyward inner layer (SIL) is compact, dense and adherentThe Earthward inner layer (EIL) is thinner than the skywardinner layer (SIL), cracked due to spallation.32) The averagethickness of the skyward and earthward rust layers are 92 and62 µm respectively. Figure 9(b) shows that the earthward rustlayers have larger number of pores and cracks through whichpenetration of deleterious species such as Cl¹ and SO2 arepossible. From the cross section morphological studies, it isseen that the earthward surface were highly influenced andaffected by corrosive agents than the skyward surface.Figure 9(c) and (d) EDAX spectra shows the presence ofchloride, sulphur and exogenous silica along with alloyingelements in the rust layers. Exogenous silica also enhancesthe corrosion protection of the steel. Further, skyward rustlayer is enriched with Cr, Mn and Ni but the earthward rustlayer is enriched with Ni only. The enrichment of Cr and Nienhances the formation of stable goethite in the skyward rustlayers when compared to the earthward rust layers.33,39)

(a) (b) (c)

Fig. 7 SEM micrograph of typical morphologies of earthward surface corrosion products of rusted weathering steel: (a) lepidocrocite,(b) goethite and (c) akaganeite.

Fig. 8 Schematic illustrations for the formation of various rust layers on weathering steel and its microstructure.

P. Dhaiveegan, N. Elangovan, T. Nishimura and N. Rajendran152

3.4 Thermal analysis of rustFigure 10(a) shows the thermogram of the rust layers. The

study reveals that weight loss of iron oxyhydroxides andoxides at different temperatures for skyward and earthwardrust layers are distinct. The significant loss of weight in thetemperature range 20­220°C is attributed to desorption of

physisorbed water molecules. subsequently, the secondweight loss occuring at 370°C is due to the acceleratedtransformation of oxyhydroxides to iron oxides, with thevarying weight loss (%) corresponding to their relativeamounts present in the skyward and earthward rust layers.12)

It is evident from Fig. 10(b). that the amount of oxy-hydroxides in comparison to iron oxides present in skywardsurface is higher than the earthward surface. Higher amountsof oxyhydroxides enhance the corrosion protection ability ofthe skyward rust layers.34)

3.5 N2 adsorption studies of rust layersN2 adsorption desorption isotherms of powedered rust

layers of SOL, SIL, EOL, EIL are shown in Fig. 11(a) and (b)respectively. From the Fig. 11(a) and (b) it is inferred that theadsorption and desorption behaviour of rust layers showstypical mesoporous material type IV isotherm which isattributed to the capillary condenstion in mesoporous rustchannels.

SOL and EOL shows the H2 type hystersis loop indicatingthe porous nature of the rust which were substantiated by theSEM results. SIL and EIL show typical H4 type hysteresisloop attributing to the less porous nature of rust which alsoaccounts for the morphological structures as revealed bySEM analysis. The hysteresis loop in the relative pressure(P/Po) range between 0.4 to 0.8 and this can be attributed tothe pores with narrow and wide sections and possibleinterconnecting channel in the rust layers.35) From, the BET

0 200 400 600 800 10000

20

40

60

80

100

Temperature, T / K

Wei

gh

t, (

%)

SOLSILEOLEIL

SOL SIL EOL EIL0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

Mas

s, m

g

Rust layer

Mass of hydroxy oxideMass of oxide

(a) (b)

Fig. 10 (a) TGA curves of powdered rust layers. (b) Relative abundance of iron oxy hydroxide and iron oxides within the various rustlayers of weathering steel.

(d)(c)

(a) (b)Outer Layer

Inner Layer

Substrate

Outer Layer

Inner Layer

Substrate

Fig. 9 Cross-sectional morphologies of rusted surface of weathering steel:(a) skyward, (b) earthward surfaces and their corresponding EDAXspectrum (c) and (d).

0.0 0.2 0.4 0.6 0.8 1.00

20

40

60

SOLSIL

Vou

lme

adso

rbed

, cm

/g3

Relative pressure, p/po

0.0 0.2 0.4 0.6 0.8 1.00

20

40

60

EILEOL

Vou

lme

adso

rbed

, cm

/g3

Relative pressure, p/po

(a) (b)

Fig. 11 N2 adsorption­desorption isotherms of powdered rust layers: (a) SOL and SIL, (b) EOL and EIL.

Weathering Steel in Industrial-Marine-Urban Environment: Field Study 153

isotherm SAwere calculated for the SIL, SOL, EIL and EOLrust layers which were found to be 107, 84, 92 and 74m2/grespectively. The SIL and EIL shows higher SA, highercorrosion resistance which is due to the compact, adherentand porous nature of rust layers. The higher SA results insmaller rust particles of the agglomerates of iron oxide andiron oxyhydroxides with finer pore structure.13) EOL andSOL show less porous which is due to the low SA resultingwith poor adherent nature. The compactness and protectiveability of the rust layers depend on the morphology of the rustparticles; smaller particles form more compact layers andsheet or plate ones make up less permeable layers by theirpreferred orientation.

The relative difference in the specific surface area of SOLand SIL can be inferred from the presence of a considerableamount of akaganeite (¢-FeOOH) which has a tunnelstructure and contributes to a high surface area and withlarge pore size. The enrichment of Ni and Cr influences thereduction of particle size in the rust and its structure andcompostion.8,36­39) The SA above 50m2/g has small poresalways filled with water by the adsorption and capillarycondensation. Further, it prevents the transfer of corrosiveagents into the pore thus leading to the high protective natureof weathering steels to atmospheric corrosion.37) From theabove results, it is clear that the SIL has a higher surface area,which contributes to the protective ability of rust layer fromatmospheric pollutants.

Figure 12(a) shows the distribution of pore volume withrespect to pore size curves calculated by BJH method fromN2 adsorption isotherm. Average pore diameter of SOL andEOL is 6 nm whereas it is 2 nm for SIL and EIL. Thedecrease in pore diameter contributes to the higher surfacearea with greater protective ability. The correlation betweenthe pore size and surface area of inner layers and outer layersare depicted in Fig. 12(b), which demonstrates the inverserelationship between SA and pore size. Reduction in porediameter also hinders the penetration of the corrosive ionssuch as Cl¹ and SO2 into the metal surface thereby acting asactive protective layers in preventing further occurance ofcorrosion processes.

4. Conclusions

Analysis of corrosion products formed on weathering steel

surfaces exposed for 3 years in the industrial-marine-urbanenvironment led to better understanding of corrosion processleading to the formation of adherent protective layer on theskyward surface when compared to earthward surface. Theresults obtained from ATR-FTIR and XRD confirmed theformation of two distinct types of layers each having uniqueoxides and oxyhydroxides on their surfaces. The formation ofgoethite in the inner layer (SIL) as stable form of productwith needle like structures was influenced by SO2, which ledto the formation of protective skyward rust layers along withother iron oxides. The BET isotherm studies confirmed thecompactness and densified nature of skyward inner rustlayers which prevents the ingress of chloride ions. Corrosionprocess with enrichment of Ni and Cr resulted in fine rustparticles in the skyward inner layer with higher surface area,thereby ensuring the corrosion protection of weathering steel.

Acknowledgement

The facilities provided by DST-FIST and UGC-DRS aregratefully acknowledged. The HR-SEM facility provided bythe SAIF-IIT, Madras is also acknowledged. The permissiongranted by Mr. P. Manikandan to expose the specimens in hisplace is gratefully acknowledged.

REFERENCES

1) Y. H. Qian, D. Niu, J. J. Xu and M. S. Li: Corros. Sci. 71 (2013) 72­77.2) D. Dasgupta: Br. Corros. J. 4 (1969) 119­121.3) M. Natesan, S. Selvaraj, T. Manickam and G. Venkatachari: Sci. Tech.

Adv. Mater. 9 (2008) 45-02.4) J. Honzák: Br. Corros. J. 8 (1973) 162­166.5) T. Nishimura: Corros. Sci. 52 (2010) 3609­3614.6) I. Suzuki: J. Electrochem. Soc. 127 (1980) 2210­2215.7) T. E. Graedel: J. Electrochem. Soc. 137 (1990) 2385­2394.8) T. Nishimura, H. Katayama, K. Noda and T. Kodama: Corros. Sci. 42

(2000) 1611­1621.9) K. Asami and K. Hashimoto: Corros. Sci. 45 (2003) 2263­2283.10) D. de la Fuente, I. Díaz, J. Simancas, B. Chico and M. Morcillo:

Corros. Sci. 53 (2011) 604­617.11) C. Genestar Juliá and C. Pons Bonafé: Thermochim. Acta 413 (2004)

185­192.12) Y. Zhao, H. Ren, H. Dai and W. Jin: Corros. Sci. 53 (2011) 1646­1658.13) T. Ishikawa, T. Yoshida, K. Kandori, T. Nakayama and S. Hara: Corros.

Sci. 49 (2007) 1468­1477.14) D. D. N. Singh, S. Yadav and J. K. Saha: Corros. Sci. 50 (2008) 93­

110.

0 5 10 15 20 25 30 35 40

0

1

2

3

4

5

6

7

8

SILEOL

dV/d

D, c

m3 /n

m

Pore Diameter, D/nm

SOL

EIL

SOL SIL EOL EIL1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

3.0Average Pore sizeSurface area

Rust layer

Ave

rage

por

e si

ze, n

m

65

70

75

80

85

90

95

100

105

110

Surface area, m2/g

(a) (b)

Fig. 12 (a) Distribution of pore volume with respect to pore size as calculated by BJH method. (b) Correlation between the pore size andspecific surface area of weathering steel rust layers.

P. Dhaiveegan, N. Elangovan, T. Nishimura and N. Rajendran154

15) Practice for conducting atmospheric corrosion of metals, G 50 ASTMWest Conshohocken, PA, USA, (1997).

16) Standard test method for determining atmospheric chloride depositionrate by wet candle method, G 140 ASTM, West Conshohocken, PA,USA, (1996).

17) Standard practice for monitoring atmospheric SO2 using the sulphationplate technique, G 91 ASTM, West Conshohocken, PA, USA, (1997).

18) ISO 9223 Corrosion of Metals and Alloys-Corrosivity of Atmospheres-Classification, (2012).

19) A. Raman, B. Kuban and A. Razvan: Corros. Sci. 32 (1991) 1295­1306.

20) M. Morcillo, D. De la Fuente, I. Díaz and H. Cano: Revis. Metall. 47(2011) 426­444.

21) P. Dhaiveegan, N. Elangovan, T. Nishimura and N. Rajendran:Electroanal. 26 (2014) 2419­2428.

22) J. P. Labbé, J. Lédion and F. Hui: Corros. Sci. 50 (2008) 1228­1234.23) F. Iwei: Br. Corros. J. 26 (1991) 209­214.24) X. H. Chen, J. H. Dong, E. H. Han and W. Ke: Corros. Eng. Sci. Tech.

42 (2007) 224­231.25) K. Chandler: Br. Corros. J. 1 (1966) 264­266.26) D. C. Cook, S. J. Oh, R. Balasubramanian and M. Yamashita:

Hyperfine Interact. 122 (1999) 59­70.

27) Y. Wan, C. Yan, J. Tan, Z.-M. Shi and C. N. Cao: J. Mater. Sci. 38(2003) 3597­3602.

28) H. Tamura: Corros. Sci. 50 (2008) 1872­1883.29) J. G. Castaño, C. A. Botero, A. H. Restrepo, E. A. Agudelo, E. Correa

and F. Echeverría: Corros. Sci. 52 (2010) 216­223.30) U. Evans: Br. Corros. J. 7 (1972) 10­14.31) T. Misawa, K. Asami, K. Hashimoto and S. Shimodaira: Corros. Sci. 14

(1974) 279­289.32) M. Taylor, P. Boden and E. Holmes: Br. Corros. J. 6 (1971) 61­65.33) Q. C. Zhang, J. S. Wu, J. J. Wang, W. L. Zheng, J. G. Chen and A. B.

Li: Mater. Chem. Phys. 77 (2003) 603­608.34) T. Kamimura, S. Hara, H. Miyuki, M. Yamashita and H. Uchida:

Corros. Sci. 48 (2006) 2799­2812.35) J. B. Condon: Surface Area and Porosity Determinations by

Physisorption: Measurements and Theory, (Elsevier, The Netherlands,2006) pp. 6­26.

36) T. Ishikawa, A. Maeda, K. Kandori and A. Tahara: Corrosion 62 (2006)559­567.

37) T. Nishimura and T. Kodama: Corros. Sci. 45 (2003) 1073­1084.38) T. Ishikawa, M. Kumagai, A. Yasukawa and K. Kandori: Corrosion 57

(2001) 346­352.39) T. Nishimura and N. Rajendran: Mater. Trans. 55 (2014) 1547­1552.

Weathering Steel in Industrial-Marine-Urban Environment: Field Study 155